The vast majority of terminally ill patients for whom life-sustaining treatments are withdrawn are "decisionally incapacitated," i.e., unable to express their own preferences about treatment. Families generally serve as representatives of these patients' values and wishes. With advances in life-sustaining medical technology and the growing number of aged and chronically ill persons, clinician-family negotiations in decision-making about limiting life supports is increasing rapidly. Yet there is little scientific knowledge about the experiences of these families. Information is needed to guide clinicians in identifying high risk families, intervening early to mediate processes and facilitate positive outcomes for families, and in shaping educational and legislative policies related to withdrawing life-sustaining treatments. Therefore, the aims of this study are to l) identify factors (patient-, family-, and clinician-related) associated with family members' responses to participation in making the decision to withdraw life-sustaining treatments; 2) compare family members and clinicians (nurses and physicians) on ethical reasoning that influenced the decision to withdraw life-sustaining treatments; and 3) describe the surrogacy role of the family related to: a) how family members take on the role of spokesperson and perceive their responsibility and authority; b) whether conflict occurred within families or between families and clinicians; c) family members' emotional and physical health; and d) family members' satisfaction over time with decision-making and the surrogacy role. Study data come from l) the charts of 30 deceased patients whose deaths followed withdrawal of life-sustaining treatments, 2) 60 of their family members, and 3) 30 of the clinicians (nurses and physicians) who cared for the patients and interacted with family members during decision- making. Data from family members and clinicians are being collected by personal interview and questionnaire within 3 weeks of the patient's death. Follow-up data are being collected only from family members by interviews and repeat questionnaires six months following the decision. Questionnaires include the Reasoning Factor Rating Scale, the Impact of Event Scale, and the RAND Health Survey. Quantitative data will be computer analyzed using descriptive and inferential statistical analyses. Narrative data from interviews are transcribed verbatim and analyzed using content analysis methods.